Allogeneic stem-cell transplantation has had limited success for patients with refractory and relapsed aggressive B-cell or T-cell lymphoma. We investigated the effect of adding rituximab to standard prophylaxis for graft-versus-host disease after transplantation and estimated overall survival when using a lymphoma-directed myeloablative conditioning regimen.

We did this randomised, open-label, phase 2 study at seven German transplantation centres. We enrolled patients with aggressive B-cell or T-cell lymphoma and primary refractory disease, early relapse (<12 months after first-line treatment), or relapse after autologous transplantation. Conditioning with fludarabine (125 mg/m²), busulfan (12 mg/kg oral or 9·6 mg/kg intravenous), and cyclophosphamide (120 mg/kg) was followed by allogeneic stem-cell transplantation. Patients were randomly assigned (1:1) to receive rituximab (375 mg/m² on days 21, 28, 35, 42, 175, 182, 189, and 196) or not. Allocation was done with a centralised computer-generated procedure; patients were stratified by histological subtype (B-cell vs T-cell lymphoma) and donor match (HLA-identical vs non-identical). Neither investigators nor patients were masked to allocation. The primary endpoints were the incidence of acute graft-versus-host disease grade 2–4 in each treatment group and overall survival at 1 year in both groups combined. All analyses were done for the intention-to-treat population. The study is registered with ClinicalTrials.gov, number NCT00785330.

Between June 16, 2004, and March 24, 2009, we screened 86 patients and enrolled 84; 42 were randomly assigned to each group. The cumulative incidence of grade 2–4 acute graft-versus-host disease was 46% (95% CI 32–62) in the rituximab group and 42% (95% CI 29–59) in the no rituximab group (hazard ratio [HR] 0·91, 95% CI 0·52–1·60; p=0·74). Overall survival at 1 year for the whole study population was 52% (95% CI 41–62). Grade 4 haematological toxic effects and grade 3 alopecia occurred in all patients. The most common non-haematological grade 5 toxic effects were pneumonia (nine in the no rituximab group vs ten in the rituximab group) and other infections (seven vs four).

The lymphoma-directed myeloablative conditioning regimen developed here is promising for patients with refractory and relapsed aggressive B-cell and T-cell lymphomas. However, the addition of rituximab did not affect the incidence of graft-versus-host disease or overall survival.

Hoffmann-La Roche, Amgen, Astellas Pharma.